Upload trainer_state.json with huggingface_hub
Browse files- trainer_state.json +1134 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,1134 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.9968,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 156,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.0128,
|
| 13 |
+
"grad_norm": 3.4233698136439568,
|
| 14 |
+
"learning_rate": 1.25e-07,
|
| 15 |
+
"loss": 0.0246,
|
| 16 |
+
"step": 1
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.0256,
|
| 20 |
+
"grad_norm": 3.728998587328461,
|
| 21 |
+
"learning_rate": 2.5e-07,
|
| 22 |
+
"loss": 0.0252,
|
| 23 |
+
"step": 2
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0384,
|
| 27 |
+
"grad_norm": 3.0616613124293135,
|
| 28 |
+
"learning_rate": 3.75e-07,
|
| 29 |
+
"loss": 0.0227,
|
| 30 |
+
"step": 3
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0512,
|
| 34 |
+
"grad_norm": 2.3123184172566016,
|
| 35 |
+
"learning_rate": 5e-07,
|
| 36 |
+
"loss": 0.0168,
|
| 37 |
+
"step": 4
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.064,
|
| 41 |
+
"grad_norm": 3.360264502123766,
|
| 42 |
+
"learning_rate": 6.249999999999999e-07,
|
| 43 |
+
"loss": 0.021,
|
| 44 |
+
"step": 5
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.0768,
|
| 48 |
+
"grad_norm": 3.3460855292395757,
|
| 49 |
+
"learning_rate": 7.5e-07,
|
| 50 |
+
"loss": 0.0229,
|
| 51 |
+
"step": 6
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.0896,
|
| 55 |
+
"grad_norm": 3.1072974219219085,
|
| 56 |
+
"learning_rate": 8.75e-07,
|
| 57 |
+
"loss": 0.0226,
|
| 58 |
+
"step": 7
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.1024,
|
| 62 |
+
"grad_norm": 4.56578073058385,
|
| 63 |
+
"learning_rate": 1e-06,
|
| 64 |
+
"loss": 0.0296,
|
| 65 |
+
"step": 8
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1152,
|
| 69 |
+
"grad_norm": 3.4123791670336443,
|
| 70 |
+
"learning_rate": 1.125e-06,
|
| 71 |
+
"loss": 0.0234,
|
| 72 |
+
"step": 9
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.128,
|
| 76 |
+
"grad_norm": 2.7894681328326816,
|
| 77 |
+
"learning_rate": 1.2499999999999999e-06,
|
| 78 |
+
"loss": 0.0203,
|
| 79 |
+
"step": 10
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.1408,
|
| 83 |
+
"grad_norm": 4.6455794479831685,
|
| 84 |
+
"learning_rate": 1.375e-06,
|
| 85 |
+
"loss": 0.0299,
|
| 86 |
+
"step": 11
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.1536,
|
| 90 |
+
"grad_norm": 3.8109983639167577,
|
| 91 |
+
"learning_rate": 1.5e-06,
|
| 92 |
+
"loss": 0.025,
|
| 93 |
+
"step": 12
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.1664,
|
| 97 |
+
"grad_norm": 4.183418083336812,
|
| 98 |
+
"learning_rate": 1.625e-06,
|
| 99 |
+
"loss": 0.0311,
|
| 100 |
+
"step": 13
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.1792,
|
| 104 |
+
"grad_norm": 4.024058901580512,
|
| 105 |
+
"learning_rate": 1.75e-06,
|
| 106 |
+
"loss": 0.024,
|
| 107 |
+
"step": 14
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.192,
|
| 111 |
+
"grad_norm": 2.9842133290060593,
|
| 112 |
+
"learning_rate": 1.8749999999999998e-06,
|
| 113 |
+
"loss": 0.0198,
|
| 114 |
+
"step": 15
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.2048,
|
| 118 |
+
"grad_norm": 4.060055578632782,
|
| 119 |
+
"learning_rate": 2e-06,
|
| 120 |
+
"loss": 0.0272,
|
| 121 |
+
"step": 16
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.2176,
|
| 125 |
+
"grad_norm": 4.646550393359002,
|
| 126 |
+
"learning_rate": 1.9997482349425066e-06,
|
| 127 |
+
"loss": 0.0216,
|
| 128 |
+
"step": 17
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2304,
|
| 132 |
+
"grad_norm": 3.9839364783795164,
|
| 133 |
+
"learning_rate": 1.9989930665413145e-06,
|
| 134 |
+
"loss": 0.0211,
|
| 135 |
+
"step": 18
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.2432,
|
| 139 |
+
"grad_norm": 4.2706791731528435,
|
| 140 |
+
"learning_rate": 1.997734875046456e-06,
|
| 141 |
+
"loss": 0.0275,
|
| 142 |
+
"step": 19
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.256,
|
| 146 |
+
"grad_norm": 4.46746121804618,
|
| 147 |
+
"learning_rate": 1.995974293995239e-06,
|
| 148 |
+
"loss": 0.0258,
|
| 149 |
+
"step": 20
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.2688,
|
| 153 |
+
"grad_norm": 4.703579744776647,
|
| 154 |
+
"learning_rate": 1.9937122098932426e-06,
|
| 155 |
+
"loss": 0.0273,
|
| 156 |
+
"step": 21
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.2816,
|
| 160 |
+
"grad_norm": 5.162187031521371,
|
| 161 |
+
"learning_rate": 1.9909497617679347e-06,
|
| 162 |
+
"loss": 0.0297,
|
| 163 |
+
"step": 22
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.2944,
|
| 167 |
+
"grad_norm": 4.08147747350908,
|
| 168 |
+
"learning_rate": 1.9876883405951377e-06,
|
| 169 |
+
"loss": 0.0241,
|
| 170 |
+
"step": 23
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.3072,
|
| 174 |
+
"grad_norm": 5.338597185310122,
|
| 175 |
+
"learning_rate": 1.9839295885986295e-06,
|
| 176 |
+
"loss": 0.0313,
|
| 177 |
+
"step": 24
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.32,
|
| 181 |
+
"grad_norm": 4.004331157501513,
|
| 182 |
+
"learning_rate": 1.9796753984232355e-06,
|
| 183 |
+
"loss": 0.0233,
|
| 184 |
+
"step": 25
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.3328,
|
| 188 |
+
"grad_norm": 5.79846158935698,
|
| 189 |
+
"learning_rate": 1.9749279121818236e-06,
|
| 190 |
+
"loss": 0.0328,
|
| 191 |
+
"step": 26
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.3456,
|
| 195 |
+
"grad_norm": 6.927108105836598,
|
| 196 |
+
"learning_rate": 1.9696895203766866e-06,
|
| 197 |
+
"loss": 0.0381,
|
| 198 |
+
"step": 27
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.3584,
|
| 202 |
+
"grad_norm": 5.251309102448725,
|
| 203 |
+
"learning_rate": 1.9639628606958534e-06,
|
| 204 |
+
"loss": 0.0282,
|
| 205 |
+
"step": 28
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.3712,
|
| 209 |
+
"grad_norm": 5.050171176886125,
|
| 210 |
+
"learning_rate": 1.9577508166849303e-06,
|
| 211 |
+
"loss": 0.0246,
|
| 212 |
+
"step": 29
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.384,
|
| 216 |
+
"grad_norm": 5.512551765259008,
|
| 217 |
+
"learning_rate": 1.9510565162951534e-06,
|
| 218 |
+
"loss": 0.03,
|
| 219 |
+
"step": 30
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.3968,
|
| 223 |
+
"grad_norm": 5.7365196994206,
|
| 224 |
+
"learning_rate": 1.9438833303083674e-06,
|
| 225 |
+
"loss": 0.0314,
|
| 226 |
+
"step": 31
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.4096,
|
| 230 |
+
"grad_norm": 5.220122317364698,
|
| 231 |
+
"learning_rate": 1.936234870639737e-06,
|
| 232 |
+
"loss": 0.0311,
|
| 233 |
+
"step": 32
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4224,
|
| 237 |
+
"grad_norm": 5.090432583516014,
|
| 238 |
+
"learning_rate": 1.928114988519039e-06,
|
| 239 |
+
"loss": 0.0289,
|
| 240 |
+
"step": 33
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.4352,
|
| 244 |
+
"grad_norm": 5.145680195739282,
|
| 245 |
+
"learning_rate": 1.9195277725514506e-06,
|
| 246 |
+
"loss": 0.0272,
|
| 247 |
+
"step": 34
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.448,
|
| 251 |
+
"grad_norm": 5.308627403286571,
|
| 252 |
+
"learning_rate": 1.9104775466588157e-06,
|
| 253 |
+
"loss": 0.0324,
|
| 254 |
+
"step": 35
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.4608,
|
| 258 |
+
"grad_norm": 6.22201023848965,
|
| 259 |
+
"learning_rate": 1.9009688679024189e-06,
|
| 260 |
+
"loss": 0.0344,
|
| 261 |
+
"step": 36
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.4736,
|
| 265 |
+
"grad_norm": 5.003597737244695,
|
| 266 |
+
"learning_rate": 1.8910065241883678e-06,
|
| 267 |
+
"loss": 0.0333,
|
| 268 |
+
"step": 37
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.4864,
|
| 272 |
+
"grad_norm": 5.6592651331248,
|
| 273 |
+
"learning_rate": 1.8805955318567379e-06,
|
| 274 |
+
"loss": 0.0315,
|
| 275 |
+
"step": 38
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.4992,
|
| 279 |
+
"grad_norm": 5.975923701038477,
|
| 280 |
+
"learning_rate": 1.8697411331556953e-06,
|
| 281 |
+
"loss": 0.0241,
|
| 282 |
+
"step": 39
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.512,
|
| 286 |
+
"grad_norm": 5.750552226599778,
|
| 287 |
+
"learning_rate": 1.858448793601866e-06,
|
| 288 |
+
"loss": 0.0329,
|
| 289 |
+
"step": 40
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.5248,
|
| 293 |
+
"grad_norm": 5.816663494605659,
|
| 294 |
+
"learning_rate": 1.8467241992282841e-06,
|
| 295 |
+
"loss": 0.0337,
|
| 296 |
+
"step": 41
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.5376,
|
| 300 |
+
"grad_norm": 5.060423336355904,
|
| 301 |
+
"learning_rate": 1.8345732537213026e-06,
|
| 302 |
+
"loss": 0.0289,
|
| 303 |
+
"step": 42
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.5504,
|
| 307 |
+
"grad_norm": 5.111705069882343,
|
| 308 |
+
"learning_rate": 1.82200207544791e-06,
|
| 309 |
+
"loss": 0.0253,
|
| 310 |
+
"step": 43
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.5632,
|
| 314 |
+
"grad_norm": 4.76340608246537,
|
| 315 |
+
"learning_rate": 1.8090169943749474e-06,
|
| 316 |
+
"loss": 0.0242,
|
| 317 |
+
"step": 44
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.576,
|
| 321 |
+
"grad_norm": 5.232632523840601,
|
| 322 |
+
"learning_rate": 1.795624548881781e-06,
|
| 323 |
+
"loss": 0.0332,
|
| 324 |
+
"step": 45
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.5888,
|
| 328 |
+
"grad_norm": 5.063499559835732,
|
| 329 |
+
"learning_rate": 1.7818314824680298e-06,
|
| 330 |
+
"loss": 0.0331,
|
| 331 |
+
"step": 46
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6016,
|
| 335 |
+
"grad_norm": 5.081572984551496,
|
| 336 |
+
"learning_rate": 1.767644740358011e-06,
|
| 337 |
+
"loss": 0.0353,
|
| 338 |
+
"step": 47
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.6144,
|
| 342 |
+
"grad_norm": 4.256457044525209,
|
| 343 |
+
"learning_rate": 1.753071466003611e-06,
|
| 344 |
+
"loss": 0.0275,
|
| 345 |
+
"step": 48
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.6272,
|
| 349 |
+
"grad_norm": 5.0457530324965925,
|
| 350 |
+
"learning_rate": 1.7381189974873407e-06,
|
| 351 |
+
"loss": 0.0345,
|
| 352 |
+
"step": 49
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.64,
|
| 356 |
+
"grad_norm": 4.222996253822678,
|
| 357 |
+
"learning_rate": 1.7227948638273915e-06,
|
| 358 |
+
"loss": 0.0258,
|
| 359 |
+
"step": 50
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.6528,
|
| 363 |
+
"grad_norm": 5.105453296008258,
|
| 364 |
+
"learning_rate": 1.7071067811865474e-06,
|
| 365 |
+
"loss": 0.0361,
|
| 366 |
+
"step": 51
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.6656,
|
| 370 |
+
"grad_norm": 4.7238042861158505,
|
| 371 |
+
"learning_rate": 1.6910626489868648e-06,
|
| 372 |
+
"loss": 0.03,
|
| 373 |
+
"step": 52
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.6784,
|
| 377 |
+
"grad_norm": 4.467100915377624,
|
| 378 |
+
"learning_rate": 1.6746705459320744e-06,
|
| 379 |
+
"loss": 0.0301,
|
| 380 |
+
"step": 53
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.6912,
|
| 384 |
+
"grad_norm": 4.199798840654239,
|
| 385 |
+
"learning_rate": 1.6579387259397126e-06,
|
| 386 |
+
"loss": 0.0272,
|
| 387 |
+
"step": 54
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.704,
|
| 391 |
+
"grad_norm": 4.253879642031892,
|
| 392 |
+
"learning_rate": 1.640875613985024e-06,
|
| 393 |
+
"loss": 0.0263,
|
| 394 |
+
"step": 55
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.7168,
|
| 398 |
+
"grad_norm": 5.478200127323976,
|
| 399 |
+
"learning_rate": 1.6234898018587336e-06,
|
| 400 |
+
"loss": 0.0369,
|
| 401 |
+
"step": 56
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.7296,
|
| 405 |
+
"grad_norm": 5.3136786533226426,
|
| 406 |
+
"learning_rate": 1.6057900438408199e-06,
|
| 407 |
+
"loss": 0.0337,
|
| 408 |
+
"step": 57
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.7424,
|
| 412 |
+
"grad_norm": 5.166894330495721,
|
| 413 |
+
"learning_rate": 1.587785252292473e-06,
|
| 414 |
+
"loss": 0.0355,
|
| 415 |
+
"step": 58
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.7552,
|
| 419 |
+
"grad_norm": 4.375886424848137,
|
| 420 |
+
"learning_rate": 1.569484493168452e-06,
|
| 421 |
+
"loss": 0.0281,
|
| 422 |
+
"step": 59
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.768,
|
| 426 |
+
"grad_norm": 6.886035234115627,
|
| 427 |
+
"learning_rate": 1.5508969814521024e-06,
|
| 428 |
+
"loss": 0.0388,
|
| 429 |
+
"step": 60
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.7808,
|
| 433 |
+
"grad_norm": 5.791644877375871,
|
| 434 |
+
"learning_rate": 1.5320320765153365e-06,
|
| 435 |
+
"loss": 0.0373,
|
| 436 |
+
"step": 61
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.7936,
|
| 440 |
+
"grad_norm": 5.096680042384012,
|
| 441 |
+
"learning_rate": 1.5128992774059062e-06,
|
| 442 |
+
"loss": 0.0344,
|
| 443 |
+
"step": 62
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.8064,
|
| 447 |
+
"grad_norm": 5.046290484669824,
|
| 448 |
+
"learning_rate": 1.4935082180643467e-06,
|
| 449 |
+
"loss": 0.0411,
|
| 450 |
+
"step": 63
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.8192,
|
| 454 |
+
"grad_norm": 5.210730266900293,
|
| 455 |
+
"learning_rate": 1.4738686624729987e-06,
|
| 456 |
+
"loss": 0.0353,
|
| 457 |
+
"step": 64
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.832,
|
| 461 |
+
"grad_norm": 4.780860746415674,
|
| 462 |
+
"learning_rate": 1.4539904997395467e-06,
|
| 463 |
+
"loss": 0.0285,
|
| 464 |
+
"step": 65
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.8448,
|
| 468 |
+
"grad_norm": 4.9827723432345765,
|
| 469 |
+
"learning_rate": 1.433883739117558e-06,
|
| 470 |
+
"loss": 0.0355,
|
| 471 |
+
"step": 66
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.8576,
|
| 475 |
+
"grad_norm": 4.566523361618775,
|
| 476 |
+
"learning_rate": 1.4135585049665206e-06,
|
| 477 |
+
"loss": 0.0229,
|
| 478 |
+
"step": 67
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 0.8704,
|
| 482 |
+
"grad_norm": 3.9442441573671534,
|
| 483 |
+
"learning_rate": 1.3930250316539235e-06,
|
| 484 |
+
"loss": 0.0251,
|
| 485 |
+
"step": 68
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 0.8832,
|
| 489 |
+
"grad_norm": 4.996881415714366,
|
| 490 |
+
"learning_rate": 1.3722936584019451e-06,
|
| 491 |
+
"loss": 0.0361,
|
| 492 |
+
"step": 69
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 0.896,
|
| 496 |
+
"grad_norm": 5.769732263820899,
|
| 497 |
+
"learning_rate": 1.3513748240813427e-06,
|
| 498 |
+
"loss": 0.0366,
|
| 499 |
+
"step": 70
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 0.9088,
|
| 503 |
+
"grad_norm": 4.573503361877803,
|
| 504 |
+
"learning_rate": 1.3302790619551672e-06,
|
| 505 |
+
"loss": 0.0272,
|
| 506 |
+
"step": 71
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.9216,
|
| 510 |
+
"grad_norm": 3.296575123346509,
|
| 511 |
+
"learning_rate": 1.3090169943749473e-06,
|
| 512 |
+
"loss": 0.023,
|
| 513 |
+
"step": 72
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 0.9344,
|
| 517 |
+
"grad_norm": 4.908588421936803,
|
| 518 |
+
"learning_rate": 1.2875993274320173e-06,
|
| 519 |
+
"loss": 0.0278,
|
| 520 |
+
"step": 73
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 0.9472,
|
| 524 |
+
"grad_norm": 3.846287378603072,
|
| 525 |
+
"learning_rate": 1.266036845566675e-06,
|
| 526 |
+
"loss": 0.0255,
|
| 527 |
+
"step": 74
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 0.96,
|
| 531 |
+
"grad_norm": 4.087209232767521,
|
| 532 |
+
"learning_rate": 1.244340406137894e-06,
|
| 533 |
+
"loss": 0.0295,
|
| 534 |
+
"step": 75
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 0.9728,
|
| 538 |
+
"grad_norm": 4.64654246748357,
|
| 539 |
+
"learning_rate": 1.2225209339563143e-06,
|
| 540 |
+
"loss": 0.0278,
|
| 541 |
+
"step": 76
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 0.9856,
|
| 545 |
+
"grad_norm": 4.507522376329423,
|
| 546 |
+
"learning_rate": 1.2005894157832728e-06,
|
| 547 |
+
"loss": 0.0319,
|
| 548 |
+
"step": 77
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.9984,
|
| 552 |
+
"grad_norm": 5.762678744609664,
|
| 553 |
+
"learning_rate": 1.1785568947986366e-06,
|
| 554 |
+
"loss": 0.0352,
|
| 555 |
+
"step": 78
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.0112,
|
| 559 |
+
"grad_norm": 3.1644394788507513,
|
| 560 |
+
"learning_rate": 1.156434465040231e-06,
|
| 561 |
+
"loss": 0.0161,
|
| 562 |
+
"step": 79
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.024,
|
| 566 |
+
"grad_norm": 3.7447721631178,
|
| 567 |
+
"learning_rate": 1.1342332658176555e-06,
|
| 568 |
+
"loss": 0.0191,
|
| 569 |
+
"step": 80
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.0368,
|
| 573 |
+
"grad_norm": 3.2322255457531637,
|
| 574 |
+
"learning_rate": 1.1119644761033077e-06,
|
| 575 |
+
"loss": 0.0124,
|
| 576 |
+
"step": 81
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.0496,
|
| 580 |
+
"grad_norm": 2.4196830627748285,
|
| 581 |
+
"learning_rate": 1.0896393089034335e-06,
|
| 582 |
+
"loss": 0.0105,
|
| 583 |
+
"step": 82
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.0624,
|
| 587 |
+
"grad_norm": 2.5920868298208872,
|
| 588 |
+
"learning_rate": 1.0672690056120398e-06,
|
| 589 |
+
"loss": 0.0144,
|
| 590 |
+
"step": 83
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.0752,
|
| 594 |
+
"grad_norm": 2.5821101862637175,
|
| 595 |
+
"learning_rate": 1.044864830350515e-06,
|
| 596 |
+
"loss": 0.0139,
|
| 597 |
+
"step": 84
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.088,
|
| 601 |
+
"grad_norm": 3.02231827873139,
|
| 602 |
+
"learning_rate": 1.022438064295805e-06,
|
| 603 |
+
"loss": 0.0135,
|
| 604 |
+
"step": 85
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.1008,
|
| 608 |
+
"grad_norm": 2.2923086363844845,
|
| 609 |
+
"learning_rate": 1e-06,
|
| 610 |
+
"loss": 0.0117,
|
| 611 |
+
"step": 86
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.1136,
|
| 615 |
+
"grad_norm": 2.8842975225041623,
|
| 616 |
+
"learning_rate": 9.77561935704195e-07,
|
| 617 |
+
"loss": 0.0115,
|
| 618 |
+
"step": 87
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.1264,
|
| 622 |
+
"grad_norm": 5.879985429015499,
|
| 623 |
+
"learning_rate": 9.551351696494853e-07,
|
| 624 |
+
"loss": 0.0119,
|
| 625 |
+
"step": 88
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.1392,
|
| 629 |
+
"grad_norm": 2.360056985536234,
|
| 630 |
+
"learning_rate": 9.327309943879603e-07,
|
| 631 |
+
"loss": 0.0096,
|
| 632 |
+
"step": 89
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.152,
|
| 636 |
+
"grad_norm": 1.9869435676758012,
|
| 637 |
+
"learning_rate": 9.103606910965665e-07,
|
| 638 |
+
"loss": 0.0079,
|
| 639 |
+
"step": 90
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.1648,
|
| 643 |
+
"grad_norm": 2.5601097067547856,
|
| 644 |
+
"learning_rate": 8.880355238966921e-07,
|
| 645 |
+
"loss": 0.0104,
|
| 646 |
+
"step": 91
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.1776,
|
| 650 |
+
"grad_norm": 2.7730596832345564,
|
| 651 |
+
"learning_rate": 8.657667341823448e-07,
|
| 652 |
+
"loss": 0.0118,
|
| 653 |
+
"step": 92
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.1904,
|
| 657 |
+
"grad_norm": 2.18163461710527,
|
| 658 |
+
"learning_rate": 8.435655349597689e-07,
|
| 659 |
+
"loss": 0.0105,
|
| 660 |
+
"step": 93
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.2032,
|
| 664 |
+
"grad_norm": 2.251464092159168,
|
| 665 |
+
"learning_rate": 8.214431052013634e-07,
|
| 666 |
+
"loss": 0.0148,
|
| 667 |
+
"step": 94
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.216,
|
| 671 |
+
"grad_norm": 2.7530295169182333,
|
| 672 |
+
"learning_rate": 7.994105842167272e-07,
|
| 673 |
+
"loss": 0.01,
|
| 674 |
+
"step": 95
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.2288000000000001,
|
| 678 |
+
"grad_norm": 2.526225960487079,
|
| 679 |
+
"learning_rate": 7.774790660436857e-07,
|
| 680 |
+
"loss": 0.0089,
|
| 681 |
+
"step": 96
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.2416,
|
| 685 |
+
"grad_norm": 2.9558648711495414,
|
| 686 |
+
"learning_rate": 7.556595938621058e-07,
|
| 687 |
+
"loss": 0.0121,
|
| 688 |
+
"step": 97
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.2544,
|
| 692 |
+
"grad_norm": 4.047664680116945,
|
| 693 |
+
"learning_rate": 7.33963154433325e-07,
|
| 694 |
+
"loss": 0.0122,
|
| 695 |
+
"step": 98
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.2671999999999999,
|
| 699 |
+
"grad_norm": 4.128501309390267,
|
| 700 |
+
"learning_rate": 7.124006725679828e-07,
|
| 701 |
+
"loss": 0.0132,
|
| 702 |
+
"step": 99
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.28,
|
| 706 |
+
"grad_norm": 3.158070174858742,
|
| 707 |
+
"learning_rate": 6.909830056250526e-07,
|
| 708 |
+
"loss": 0.0105,
|
| 709 |
+
"step": 100
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.2928,
|
| 713 |
+
"grad_norm": 2.330265912872272,
|
| 714 |
+
"learning_rate": 6.697209380448332e-07,
|
| 715 |
+
"loss": 0.0101,
|
| 716 |
+
"step": 101
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.3056,
|
| 720 |
+
"grad_norm": 4.9757360072405445,
|
| 721 |
+
"learning_rate": 6.486251759186572e-07,
|
| 722 |
+
"loss": 0.0179,
|
| 723 |
+
"step": 102
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.3184,
|
| 727 |
+
"grad_norm": 3.8743181605977743,
|
| 728 |
+
"learning_rate": 6.277063415980548e-07,
|
| 729 |
+
"loss": 0.0129,
|
| 730 |
+
"step": 103
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.3312,
|
| 734 |
+
"grad_norm": 3.263723675313523,
|
| 735 |
+
"learning_rate": 6.069749683460764e-07,
|
| 736 |
+
"loss": 0.0111,
|
| 737 |
+
"step": 104
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.3439999999999999,
|
| 741 |
+
"grad_norm": 3.1218175870205584,
|
| 742 |
+
"learning_rate": 5.864414950334795e-07,
|
| 743 |
+
"loss": 0.0119,
|
| 744 |
+
"step": 105
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.3568,
|
| 748 |
+
"grad_norm": 3.717962785205817,
|
| 749 |
+
"learning_rate": 5.661162608824419e-07,
|
| 750 |
+
"loss": 0.0115,
|
| 751 |
+
"step": 106
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.3696,
|
| 755 |
+
"grad_norm": 3.650556269715187,
|
| 756 |
+
"learning_rate": 5.460095002604532e-07,
|
| 757 |
+
"loss": 0.0123,
|
| 758 |
+
"step": 107
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.3824,
|
| 762 |
+
"grad_norm": 3.2197493950580296,
|
| 763 |
+
"learning_rate": 5.261313375270013e-07,
|
| 764 |
+
"loss": 0.0137,
|
| 765 |
+
"step": 108
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.3952,
|
| 769 |
+
"grad_norm": 3.365111064634147,
|
| 770 |
+
"learning_rate": 5.064917819356531e-07,
|
| 771 |
+
"loss": 0.0111,
|
| 772 |
+
"step": 109
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.408,
|
| 776 |
+
"grad_norm": 4.710390460257863,
|
| 777 |
+
"learning_rate": 4.871007225940939e-07,
|
| 778 |
+
"loss": 0.0129,
|
| 779 |
+
"step": 110
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.4208,
|
| 783 |
+
"grad_norm": 2.76927802183368,
|
| 784 |
+
"learning_rate": 4.6796792348466353e-07,
|
| 785 |
+
"loss": 0.013,
|
| 786 |
+
"step": 111
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.4336,
|
| 790 |
+
"grad_norm": 3.2171761582689915,
|
| 791 |
+
"learning_rate": 4.4910301854789755e-07,
|
| 792 |
+
"loss": 0.0114,
|
| 793 |
+
"step": 112
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.4464000000000001,
|
| 797 |
+
"grad_norm": 2.7947744875678096,
|
| 798 |
+
"learning_rate": 4.3051550683154804e-07,
|
| 799 |
+
"loss": 0.0113,
|
| 800 |
+
"step": 113
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.4592,
|
| 804 |
+
"grad_norm": 2.4585140708787043,
|
| 805 |
+
"learning_rate": 4.1221474770752696e-07,
|
| 806 |
+
"loss": 0.0103,
|
| 807 |
+
"step": 114
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.472,
|
| 811 |
+
"grad_norm": 2.8113536653109965,
|
| 812 |
+
"learning_rate": 3.942099561591802e-07,
|
| 813 |
+
"loss": 0.0106,
|
| 814 |
+
"step": 115
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.4848,
|
| 818 |
+
"grad_norm": 3.6398945452240055,
|
| 819 |
+
"learning_rate": 3.765101981412665e-07,
|
| 820 |
+
"loss": 0.0127,
|
| 821 |
+
"step": 116
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.4976,
|
| 825 |
+
"grad_norm": 2.9485443643029607,
|
| 826 |
+
"learning_rate": 3.5912438601497584e-07,
|
| 827 |
+
"loss": 0.009,
|
| 828 |
+
"step": 117
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.5104,
|
| 832 |
+
"grad_norm": 2.9984190637681096,
|
| 833 |
+
"learning_rate": 3.420612740602874e-07,
|
| 834 |
+
"loss": 0.0093,
|
| 835 |
+
"step": 118
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.5232,
|
| 839 |
+
"grad_norm": 2.8046736646132744,
|
| 840 |
+
"learning_rate": 3.253294540679257e-07,
|
| 841 |
+
"loss": 0.0094,
|
| 842 |
+
"step": 119
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.536,
|
| 846 |
+
"grad_norm": 2.9182942187963605,
|
| 847 |
+
"learning_rate": 3.0893735101313535e-07,
|
| 848 |
+
"loss": 0.0101,
|
| 849 |
+
"step": 120
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.5488,
|
| 853 |
+
"grad_norm": 4.061738080588852,
|
| 854 |
+
"learning_rate": 2.9289321881345254e-07,
|
| 855 |
+
"loss": 0.0148,
|
| 856 |
+
"step": 121
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.5615999999999999,
|
| 860 |
+
"grad_norm": 2.6181262527250064,
|
| 861 |
+
"learning_rate": 2.7720513617260855e-07,
|
| 862 |
+
"loss": 0.0108,
|
| 863 |
+
"step": 122
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.5744,
|
| 867 |
+
"grad_norm": 2.814162128761838,
|
| 868 |
+
"learning_rate": 2.6188100251265943e-07,
|
| 869 |
+
"loss": 0.0085,
|
| 870 |
+
"step": 123
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.5872000000000002,
|
| 874 |
+
"grad_norm": 2.978469400791401,
|
| 875 |
+
"learning_rate": 2.4692853399638913e-07,
|
| 876 |
+
"loss": 0.0123,
|
| 877 |
+
"step": 124
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.6,
|
| 881 |
+
"grad_norm": 1.9752826330171183,
|
| 882 |
+
"learning_rate": 2.3235525964198888e-07,
|
| 883 |
+
"loss": 0.0091,
|
| 884 |
+
"step": 125
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.6128,
|
| 888 |
+
"grad_norm": 2.6883643677418623,
|
| 889 |
+
"learning_rate": 2.181685175319702e-07,
|
| 890 |
+
"loss": 0.0097,
|
| 891 |
+
"step": 126
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.6256,
|
| 895 |
+
"grad_norm": 2.9489506229688884,
|
| 896 |
+
"learning_rate": 2.043754511182191e-07,
|
| 897 |
+
"loss": 0.0079,
|
| 898 |
+
"step": 127
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.6383999999999999,
|
| 902 |
+
"grad_norm": 1.9593044592383062,
|
| 903 |
+
"learning_rate": 1.9098300562505264e-07,
|
| 904 |
+
"loss": 0.0081,
|
| 905 |
+
"step": 128
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.6512,
|
| 909 |
+
"grad_norm": 1.952957689869861,
|
| 910 |
+
"learning_rate": 1.7799792455209016e-07,
|
| 911 |
+
"loss": 0.0082,
|
| 912 |
+
"step": 129
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.6640000000000001,
|
| 916 |
+
"grad_norm": 2.38161666441156,
|
| 917 |
+
"learning_rate": 1.6542674627869734e-07,
|
| 918 |
+
"loss": 0.0094,
|
| 919 |
+
"step": 130
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.6768,
|
| 923 |
+
"grad_norm": 2.164377402243927,
|
| 924 |
+
"learning_rate": 1.5327580077171588e-07,
|
| 925 |
+
"loss": 0.0097,
|
| 926 |
+
"step": 131
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.6896,
|
| 930 |
+
"grad_norm": 3.518297727779028,
|
| 931 |
+
"learning_rate": 1.415512063981339e-07,
|
| 932 |
+
"loss": 0.0089,
|
| 933 |
+
"step": 132
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.7024,
|
| 937 |
+
"grad_norm": 3.697074996731219,
|
| 938 |
+
"learning_rate": 1.3025886684430465e-07,
|
| 939 |
+
"loss": 0.0078,
|
| 940 |
+
"step": 133
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.7151999999999998,
|
| 944 |
+
"grad_norm": 4.314207988612093,
|
| 945 |
+
"learning_rate": 1.19404468143262e-07,
|
| 946 |
+
"loss": 0.0119,
|
| 947 |
+
"step": 134
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.728,
|
| 951 |
+
"grad_norm": 2.9026476752647414,
|
| 952 |
+
"learning_rate": 1.089934758116322e-07,
|
| 953 |
+
"loss": 0.0138,
|
| 954 |
+
"step": 135
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 1.7408000000000001,
|
| 958 |
+
"grad_norm": 2.502384977722473,
|
| 959 |
+
"learning_rate": 9.903113209758096e-08,
|
| 960 |
+
"loss": 0.0112,
|
| 961 |
+
"step": 136
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 1.7536,
|
| 965 |
+
"grad_norm": 1.7807221172577514,
|
| 966 |
+
"learning_rate": 8.952245334118413e-08,
|
| 967 |
+
"loss": 0.0077,
|
| 968 |
+
"step": 137
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 1.7664,
|
| 972 |
+
"grad_norm": 2.727436513377534,
|
| 973 |
+
"learning_rate": 8.047222744854942e-08,
|
| 974 |
+
"loss": 0.0096,
|
| 975 |
+
"step": 138
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 1.7792,
|
| 979 |
+
"grad_norm": 1.6157555666433816,
|
| 980 |
+
"learning_rate": 7.188501148096116e-08,
|
| 981 |
+
"loss": 0.007,
|
| 982 |
+
"step": 139
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 1.792,
|
| 986 |
+
"grad_norm": 1.5949515973033697,
|
| 987 |
+
"learning_rate": 6.376512936026279e-08,
|
| 988 |
+
"loss": 0.0062,
|
| 989 |
+
"step": 140
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 1.8048,
|
| 993 |
+
"grad_norm": 2.6399637194756718,
|
| 994 |
+
"learning_rate": 5.611666969163242e-08,
|
| 995 |
+
"loss": 0.0095,
|
| 996 |
+
"step": 141
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 1.8176,
|
| 1000 |
+
"grad_norm": 3.0621545226623907,
|
| 1001 |
+
"learning_rate": 4.8943483704846465e-08,
|
| 1002 |
+
"loss": 0.0132,
|
| 1003 |
+
"step": 142
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 1.8304,
|
| 1007 |
+
"grad_norm": 1.9600404784878551,
|
| 1008 |
+
"learning_rate": 4.224918331506955e-08,
|
| 1009 |
+
"loss": 0.0095,
|
| 1010 |
+
"step": 143
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 1.8432,
|
| 1014 |
+
"grad_norm": 1.963746557604649,
|
| 1015 |
+
"learning_rate": 3.6037139304146756e-08,
|
| 1016 |
+
"loss": 0.0099,
|
| 1017 |
+
"step": 144
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 1.8559999999999999,
|
| 1021 |
+
"grad_norm": 4.527295989762448,
|
| 1022 |
+
"learning_rate": 3.0310479623313125e-08,
|
| 1023 |
+
"loss": 0.0144,
|
| 1024 |
+
"step": 145
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 1.8688,
|
| 1028 |
+
"grad_norm": 2.6834067438548166,
|
| 1029 |
+
"learning_rate": 2.507208781817638e-08,
|
| 1030 |
+
"loss": 0.012,
|
| 1031 |
+
"step": 146
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 1.8816000000000002,
|
| 1035 |
+
"grad_norm": 1.8417342752412211,
|
| 1036 |
+
"learning_rate": 2.032460157676452e-08,
|
| 1037 |
+
"loss": 0.0077,
|
| 1038 |
+
"step": 147
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 1.8944,
|
| 1042 |
+
"grad_norm": 2.99373797619176,
|
| 1043 |
+
"learning_rate": 1.607041140137033e-08,
|
| 1044 |
+
"loss": 0.0115,
|
| 1045 |
+
"step": 148
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 1.9072,
|
| 1049 |
+
"grad_norm": 1.1214911188004222,
|
| 1050 |
+
"learning_rate": 1.231165940486234e-08,
|
| 1051 |
+
"loss": 0.0062,
|
| 1052 |
+
"step": 149
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 1.92,
|
| 1056 |
+
"grad_norm": 2.116183152950272,
|
| 1057 |
+
"learning_rate": 9.050238232065299e-09,
|
| 1058 |
+
"loss": 0.0097,
|
| 1059 |
+
"step": 150
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 1.9327999999999999,
|
| 1063 |
+
"grad_norm": 1.8359832453089462,
|
| 1064 |
+
"learning_rate": 6.2877901067573955e-09,
|
| 1065 |
+
"loss": 0.0069,
|
| 1066 |
+
"step": 151
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 1.9456,
|
| 1070 |
+
"grad_norm": 2.0445129237685773,
|
| 1071 |
+
"learning_rate": 4.025706004760931e-09,
|
| 1072 |
+
"loss": 0.0086,
|
| 1073 |
+
"step": 152
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 1.9584000000000001,
|
| 1077 |
+
"grad_norm": 1.4095794512624829,
|
| 1078 |
+
"learning_rate": 2.2651249535439177e-09,
|
| 1079 |
+
"loss": 0.0053,
|
| 1080 |
+
"step": 153
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 1.9712,
|
| 1084 |
+
"grad_norm": 1.4980650303315703,
|
| 1085 |
+
"learning_rate": 1.0069334586854105e-09,
|
| 1086 |
+
"loss": 0.0068,
|
| 1087 |
+
"step": 154
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 1.984,
|
| 1091 |
+
"grad_norm": 1.6817715164958336,
|
| 1092 |
+
"learning_rate": 2.517650574934693e-10,
|
| 1093 |
+
"loss": 0.0087,
|
| 1094 |
+
"step": 155
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 1.9968,
|
| 1098 |
+
"grad_norm": 5.327422033689792,
|
| 1099 |
+
"learning_rate": 0.0,
|
| 1100 |
+
"loss": 0.0163,
|
| 1101 |
+
"step": 156
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 1.9968,
|
| 1105 |
+
"step": 156,
|
| 1106 |
+
"total_flos": 138561371045888.0,
|
| 1107 |
+
"train_loss": 0.019949143110678937,
|
| 1108 |
+
"train_runtime": 6111.638,
|
| 1109 |
+
"train_samples_per_second": 6.545,
|
| 1110 |
+
"train_steps_per_second": 0.026
|
| 1111 |
+
}
|
| 1112 |
+
],
|
| 1113 |
+
"logging_steps": 1,
|
| 1114 |
+
"max_steps": 156,
|
| 1115 |
+
"num_input_tokens_seen": 0,
|
| 1116 |
+
"num_train_epochs": 2,
|
| 1117 |
+
"save_steps": 100,
|
| 1118 |
+
"stateful_callbacks": {
|
| 1119 |
+
"TrainerControl": {
|
| 1120 |
+
"args": {
|
| 1121 |
+
"should_epoch_stop": false,
|
| 1122 |
+
"should_evaluate": false,
|
| 1123 |
+
"should_log": false,
|
| 1124 |
+
"should_save": true,
|
| 1125 |
+
"should_training_stop": true
|
| 1126 |
+
},
|
| 1127 |
+
"attributes": {}
|
| 1128 |
+
}
|
| 1129 |
+
},
|
| 1130 |
+
"total_flos": 138561371045888.0,
|
| 1131 |
+
"train_batch_size": 2,
|
| 1132 |
+
"trial_name": null,
|
| 1133 |
+
"trial_params": null
|
| 1134 |
+
}
|